v6.2.2

pbakaus/impeccablev6.2.2Sep 17, 2026by adamnsandle

Upgrading?

AI Summary

Introduces an offline ONNX sequence model to release the Global Interpreter Lock (GIL), improving performance for 16kHz audio processing.

Key Highlights

  • New offline ONNX sequence model added to release GIL.
  • Currently supports 16 kHz audio only.
  • Separate `silero_vad_16k_sequence.onnx` model file.
  • New function `get_speech_timestamps_sequence` available.

New Features

  • New `get_speech_timestamps_sequence` function.
  • New `silero_vad_16k_sequence.onnx` model file.

Full Release Notes

Added offline ONNX sequence to release GIL- #795
- 16 kHz only for now
- separate `silero_vad_16k_sequence.onnx` model

```python3
from silero_vad import load_silero_vad, get_speech_timestamps_sequence, read_audio

model = load_silero_vad(sequence=True)          # loads silero_vad_16k_sequence.onnx
audio = read_audio("audio.wav", sampling_rate=16000)
ts = get_speech_timestamps_sequence(audio, model, return_seconds=True)
```

Related Repositories